MSD Hertford Road Hoddesdon , Hertfordshire

MSD. Registered Office Hertford Road, Hoddesdon, Hertfordshire EN11 9BU Registered in England

No. 820771

1

Kate Moore

Technology Appraisals Project Manager

National Institute for Health and Clinical Excellence

Level 1A, City Tower, Piccadilly Plaza

Manchester M1 4BD

12th

November 2010

Dear Ms. Moore:

RE: GOLIMUMAB FOR THE TREATMENT OF RHEUMATOID

ARTHRITIS ("RA") AFTER THE FAILURE OF PREVIOUS DISEASE-

MODIFYING ANTI-RHEUMATIC DRUGS ("DMARDS") – COMMENTS

ON THE APPRAISAL CONSULTATION DOCUMENT ("ACD")

Schering-Plough Limited, which is now part of MSD ("MSD"), welcomes the

opportunity to comment on the ACD, which sets out the Appraisal Committee’s

("the Committee") recommendations on golimumab for the treatment of RA.

We are disappointed that the Committee, having reviewed all of the evidence as

well as hearing from stakeholders, has not felt able to recommend golimumab for

the treatment of patients suffering from RA in a DMARD experienced or TNFα

inhibitor experienced population.

There is a role for golimumab, based on significantly reduced injection frequency

and fewer injection site reactions, reducing pain and discomfort for the patient that

translates into a better quality of life. In addition, it provides physicians with a

further treatment option to enable the more effective management of RA. This was

clearly articulated by the patient representatives and the clinical experts in both

submissions to, and depositions at the Committee meeting.

MSD believes that the original submission with the addition of the analyses

provided below demonstrates that golimumab is both clinically efficacious and

cost-effective for use in the treatment of DMARD experienced and TNFα

inhibitor experienced patients with RA.

The response to the Committee request for additional analyses in section 1.4,

MSD follows:

MSD

Hertford Road

Hoddesdon , Hertfordshire

EN11 9BU, UK

Telephone +44 (0)1992 452644

Facsimile +44 (0)1992 468175


No. 820771

2

1. 'incorporation of ACR70 data in the economic model'

MSD has incorporated the ACR70 data into the 'methotrexate experienced

patients' model as requested.

The addition of the ACR70 data increases all of the ICERs and reverses the

relative positions of certolizumab and infliximab (Tables 1 and 2 below).

Table 1: Original base case analysis

Total Costs (£)

Total QALYs ICER (£) vs. baseline (methotrexate)

Methotrexate 35,869 4.569 -

Infliximab 69,899 5.651 31,451

Certolizumab 73,571 5.768 31,445

Adalimumab 66,875 5.792 25,352

Golimumab 67,747 5.827 25,340

Etanercept 74,208 6.133 24,513

Table 2: Revised base case analysis incorporating ACR70 data

Total Costs (£)



Certolizumab 77,348 6.252 39,529

Infliximab 78,527 6.447 34,024

Adalimumab 70,514 6.323 30,451

Golimumab 74,201 6.554 27,862

Etanercept 83,472 6.900 26,795

The addition of the ACR70 data and changing the HAQ progression rate from

0.09 to 0.06 is also presented for completeness (Table 3 below).

Table 3: Revised base case analysis with the inclusion of ACR70 data and HAQ progression

changed to 0.06

This revised analysis places all interventions comfortably above an ICER

threshold of £30,000 and leaves the relative positioning as for Table 2.

Total Costs (£)



Certolizumab 77,,348 6.946 43,721

Infliximab 78,527 7.070 39,564

Golimumab 70,514 6.930 36,728

Etanercept 83,472 7.387 33,884

Adalimumab 74,201 7.143 32,955


No. 820771

3

The comment in the ERG regarding the base case results presented by the

manufacturer (p.121):

"…shows that infliximab and certolizumab are both dominated by golimumab,

because golimumab is more effective and less costly. The remaining strategies

all have very similar ICERs when compared to methotrexate, at around

£25,000. The incremental analysis shows that adalimumab and golimumab are

both extendedly dominated by etanercept. Etanercept generates the most

QALYs of any strategy, but at a lower cost per QALY ratio. The full

incremental analysis shows that etanercept is the optimal strategy, with an

ICER of £21,000 compared to golimumab. The analysis also shows that

golimumab is a cost-effective strategy when compared to infliximab and

certolizumab, which are already recommended by NICE."

From the point of view of relative efficacy the text essentially still applies to the

revised analysis; as does the concluding comment regarding "golimumab being a

cost-effective strategy when compared to infliximab and certolizumab, which are

already recommended by NICE."

2. 'provision of SF-36 data from the GO-FORWARD and GO-AFTER

trials…'

MSD has obtained the SF-36 data for weeks 14 and 24 from the GO-FORWARD

study (see Appendix 1).

MSD is unable to provide SF-36 data from the GO-AFTER study given that it did

not form part of the study protocol and was therefore not collected.

3. '...and a sensitivity analysis in which these data are included in the

economic model using SF-6D and/or mapping approaches to EQ-5D'

MSD has now been able to obtain individual patient level data to inform our

response to the ACD.

Individual patient level data has been mapped to SF-6D using the algorithm

developed by Sheffield University (http://www.shef.ac.uk/scharr/sections/heds/mvh/sf-

6d/revisions.html). We present both the parametric and posterior estimates. ACR70

data was not available within the provided dataset and therefore the ACR 70

values are assumed to be the same as for ACR50 (this is likely to be conservative).

We have provided the estimates in appendix 2.

The sensitivity analysis was run in the model after it was updated with the ACR70

results for all other TNFα Inhibitors. Please note that although this analysis may

(with considerable caution) be compared against Tables 1 and 2 it should not be

compared with the results in Table 3. The results are presented below in Tables 4

and 5.

http://www.shef.ac.uk/scharr/sections/heds/mvh/sf-6d/revisions.html

http://www.shef.ac.uk/scharr/sections/heds/mvh/sf-6d/revisions.html


No. 820771

4

Table 4: Total costs and QALYs for golimumab and methotrexate in one way sensitivity

analysis SF-36 to SF-6D (parametric)

Table 5: Total costs and QALYs for golimumab and methotrexate in one way sensitivity

analysis SF-36 to SF-6D (posterior)

There are a number of caveats attached to this analysis:

Inputting the mapped SF-6D data not only modifies the golimumab total

costs and QALYs gained but also modifies those for methotrexate as can

be seen when comparing against the comparable values in Table 2.

We have not been able to perform the same mapping for the other TNFα

Inhibitors

There is significant concern regarding the ability of SF-36 and therefore

SF-6D to accurately capture/reflect the utility associated with patient

reported outcomes in patients with rheumatoid arthritis (Adams et al, 2010;

Bansback et al, 2007; Hurst et al, 1997; Ruta et al, 1998; Scott, D., &

Garrood, T., 2000).

SF-6D is consistently seen to underestimate utility in relation to EQ-5D

(Marra et al, 2004).

Given the caveats the estimates do provide face validity to the derived values used

in the original submission as evidenced in Table 6 below.

Table 6: ICERs derived from Sheffield algorithm SF-6D mapped estimates.

Total Costs (£) Total QALYs ICER (£) versus baseline (Methotrexate)

methotrexate 36,485 7.214 -

Golimumab 68,824 8.085 37,129

Total Costs (£) Total QALYs ICER (£) versus baseline (Methotrexate)

methotrexate 36,716 7.196 -

Golimumab 69,054 8.066 37,170

Total Costs (£) Total QALYs

ICER (£) vs. baseline (methotrexate)


Certolizumab 75,499 8.062 46,913

Infliximab 76,678 8.207 41,174

Golimumab 68,666 8.086 37,647

Etanercept 81,627 8.501 35,557

Adalimumab 72,352 8.278 34,277


No. 820771

5

3. 'data including the proportion of people who will receive 100 mg

golimumab (that is, people who weigh more than 100 kg and whose disease

has not responded after three or four doses) and inclusion of this proportion

in the economic model.'

The Committee may or may not be aware that Committee C, their counterpart

reviewing golimumab for use in patients with Psoriatic Arthritis (PsA), had

expressed concerns regarding potential dose escalation to 100mg and the

subsequent effect on cost-effectiveness.

MSD took the decision to propose a patient access scheme to address the concerns

of Committee C.

MSD does not believe that dose-escalation will occur for any of the three

indications for which golimumab is currently licensed. This view is supported by

the clinical experts however the scheme (outline details below) has been put in

place to ensure that the NHS will not bear any additional cost should dose

escalation occur, and in such a way as to essentially remove any associated

administrative burden from the NHS.

Patient Access Scheme

MSD has submitted a request to the Department of Health for consideration of a

Patient Access Scheme for golimumab that will apply across all three licensed

indications (PsA, RA and ankylosing spondylitis (AS)). This will have the effect

of 'flat pricing' golimumab irrespective of whether the patient is prescribed 50mg

pcm (1 x 50mg auto injector) or 100mg pcm (2 x 50mg auto injector). I.e.

irrespective of whether a patient weighing >100kg is treated with 50mg pcm or

100mg pcm, the cost to the NHS will be as for a dose of 50mg pcm.

The scheme has been designed to minimise any impact on the NHS by placing the

administrative burden on the wholesaler (Medco) when patients are prescribed

golimumab at a total dose of 100mg pcm outside of the hospital setting. This

means that patients who are prescribed the 100mg dose will receive it at the price

of the 50mg dose, with the only action required being to request the appropriate

dose on the prescription request. For the small number of patients who may be

treated from hospital stocks (unlikely to occur in reality), MSD will work with the

wholesaler to simplify any audit/reconciliation required.

It should also be noted that prescribing a dose of 100mg (2 x 50mg auto injector)

can only occur for patients who weigh more than 100kg and are described as

inadequately responding to a 50mg dose given the prescribing metrics between the

provider and Medco.

We have not re-run the cost-effectiveness analyses, given that we believe there

will be only rare use of a total dose of 100mg pcm. The PAS would absorb any

additional cost should dose escalation occur.


No. 820771

6

We would also note that we are aware that the respective timelines of NICE and

PASLU militate against presenting the Committee with an approved scheme for

the November 25th

meeting.

It is possible that the Committee conclude in its final guidance that there is no

compelling evidence to support dose escalation to a total of 100mg pcm and

therefore does not recommend clinicians to do so from the perspective of cost-

effectiveness. Given that there is no direct clinical data supporting dose escalation

for patients weighing >100kg, MSD will not be advocating dose escalation and

would thus not be marketing golimumab at odds with such a recommendation

were it to be included in the final guidance.

4. 'a sensitivity analysis in which disease progression on palliative treatment

is reflected as an increase in HAQ score of 0.06 per year'

We have re-run the analysis after modifying the HAQ progression for palliative

treatment to 0.06 per year (Table 7 below).

Table 7: Original base case analysis with palliative treatment HAQ progression of 0.06 pa

This has the effect of increasing the ICERs for all five TNFα Inhibitors and does

not modify their original relative relationship. An analysis incorporating this

change plus the addition of ACR70 has been provided above (Table 3, p.2).

5. 'cost-effectiveness results for the population in 1.3 for golimumab

compared with adalimumab, etanercept, infliximab, abatacept and

tocilizumab.'

Given the (lack of) availability of comparator data MSD has conducted an

analysis comparing golimumab to only tocilizumab. The results are presented

below in table 8.

Total Costs (£)



Infliximab 76,659 6.430 39,142

Certolizumab 77,296 6.538 35,774

Adalimumab 71,467 6.550 29,968

Golimumab 73,082 6.608 29,860

Etanercept 79,759 6.863 29,057


No. 820771

7

Table 8: Golimumab compared with tocilizumab in TNFα Inhibitor experienced patients.

CONCLUSION

MSD is confident that the analyses provided in this response, in addition to our

original submission, provide the Committee with the required information to

modify provisional recommendations 1.1 and 1.3 to recommend golimumab for

use in line with TNFα Inhibitor guidance in TA130 and TA195 respectively.

The decision to do so will result in an enhancement to the physician's

armamentarium as well as providing a valuable option for patients who need

flexibility in their treatment regimen to maintain a reasonable quality of life.

MSD would also argue that the additional analysis provided above comparing the

use of golimumab versus tocilizumab in patients who have received a previous

TNFα Inhibitor should lead to a re-consideration of point 1.2 of the provisional

recommendation.

MSD will cooperate in the provision of any other information or analyses that the

Committee might wish to review.

Sincerely,

xxxxxxxxxxxxx

xxxxxxxxxxxxxxxxxxxxxxxxxxxxx

MSD

Technologies Total Costs (£) Total QALYs ICER (£) versus baseline (Methotrexate)

Incremental analysis

methotrexate 37,134 3.849 - -

Tocilizumab 51,207 4.210 38,983 38,983

Golimumab 53,519 4.361 32,002 17,927

Rituximab 53,530 4.514 24,656 72


No. 820771

8

Appendix 1

Summary of SF-36 physical and mental component summary scores at baseline;

randomized subjects


No. 820771

9

Summary of norm-based scores of SF-36 scales at baseline; randomized subjects


No. 820771

10


No. 820771

11

Summary of change from baseline in SF-36 physical component summary scores

at Week 14 and Week 24; randomized subjects


No. 820771

12

Summary of change from baseline in norm-based scores of SF-36 scales at Week

14 and Week 24; randomized subjects


No. 820771

13


No. 820771

14


No. 820771

15

Appendix 2

Utilities - Final.xls


No. 820771

16

References

Adams, R., Walsh, C., Veale, D., Bresnihan, B., FitzGerald, O., & Barry, M. (2010) 'Understanding the

Relationship between the EQ-5D, SF-6D, HAQ and Disease Activity in Inflammatory Arthritis',

Pharmacoeconomics, Vol. 28, No. 6, Pp. 477-487

Bansback, N., Marra, C., Tsuchiya, A., Anis, A., Guh, D., Hammond, T., & Braizer, J. (2007) 'Using the

Health Assessment Questionnaire to Estimate Preference-Based Single Indices in Patients with

Rheumatoid Arthritis', Arthritis & Rheumatism, Vol. 57, No. 6, Pp. 963-971

Hurst, N., Kind, P., Ruta, D., Hunter, M., Stubbings, A. (1997) 'Measuring Health-Related Quality of Life

in Rheumatoid Arthritis: Validity, Responsiveness and Relaibility of EuroQol (EQ-5D)', British Journal of

Rheumatology, Vol. 36, No. 5, Pp. 551-559

Marra, C., Esdaile, J., Guh, D., Kopec, J., Braizer, J., Koehler, B., Chalmers, A., & Anis, A. (2004) 'A

Comparison of Four Indirect Methods of Assessing Utility Values in Rheumatoid Arthritis', Medical

Care, Vol. 42, No. 11, Pp. 1125-1131

Ruta, D., Hurst, N., Kind, P., Hunter, M., & Stubbings, A. (1998) 'Measuring health status in British

patients with rheumatoid arthritis: reliability, validity and responsiveness of the short form 36-item

health survey (SF-36)', Rheumatology, Vol.37, No. 4, Pp. 425-436

Scott, D., & Garrood, T. (2000) 'Quality of Life Measures: Use and Abuse', Bailliere's Clinical

Rheumatology, Vol. 14, No. 4, Pp. 663-687

Documents

MSD Hertford Road Hoddesdon , Hertfordshire